Benzene is a human and rodent carcinogen. Public health concern over benzene is warranted because it is ubiquitous in air, water, soil, and, thus, in food and beverages. Exposure to benzene is associated with depression of blood forming elements leading to anaplastic anemia, followed by myelodysplastic syndrome (MDS) and, ultimately, to leukemia or lymphoma. Inhalation of benzene results in a slower rate of delivery and a greater internal dose than other routes of exposure. At equivalent oral dose (mg/kg), more benzene is expired unmetabolized after administration by the oral route (60%) than by inhalation (14%). Mice may also produce a greater variety of benzene metabolites by inhalation. Most critical to selection of dose for experimental studies is to know the exposure level that becomes saturating to pathways of detoxification. For inhalation exposure this is 200 ppm (6 h TWA. Between 5 and 50 ppm benzene (6 h TWA, there is no significant difference between urinary metabolites. Available data also suggests that the ratio of hydroquinone or muconic acid to phenol ratio after inhalation exposure to 50 ppm (6 h TWA) is closer between mice and humans than either rats or Cynomolgus monkeys. This is critical because these may be the most toxic benzene intermediates. Thus, an exposure model must take into account saturation of benzene metabolism pathways, available data on exposures and high-affinity, low capacity pathways of metabolism that result in the most hematotoxic intermediates. We have been able to show that p53 deficient mice exposed to low levels (100 ppm, 6 h TWA) develop thymic lymphomas rapidly whereas mice exposed to higher levels (200 ppm, 6 h TWA) less rapidly and a significantly decreased incidence. In these studies, benzene induced lymphomas in the p53 deficient mice were: 1) clonal (T-cell receptor rearrangements were common), 2) showed a pattern of loss or deletions in chromosome 11 carrying the p53 wildtype allele different from sporadic lymphomas, and 3) showed a pattern of dysregulation of critical genes in both the p53 and Rb pathways that affected cell cycle control and population growth and apoptosis. We are continuing to investigate the effect of benzene dose and dose rate exposure by inhalation or drinking water. Using low doses and reduced frequency of exposure, metabolic pathways in the hematopoietic stem cell compartment that show high affinity for oxidation of benzene to mutagenic metabolites are targeted. At high doses and rates, these data suggest that conjugated metabolites are efficiently excreted. More research is required to investigate these phenomena.